torchmil/Camelyon16_MIL
收藏Hugging Face2025-10-20 更新2025-11-01 收录
下载链接：
https://hf-mirror.com/datasets/torchmil/Camelyon16_MIL
下载链接
链接失效反馈官方服务：
资源简介：
CAMELYON16MIL数据集是针对多实例学习（MIL）的CAMELYON16数据集的改编版本。该数据集包含经过预处理的淋巴节点切片的全景图像（WSIs），其中提取了相关的斑块并计算了每个斑块的特征。如果一个斑块的超过50%的像素被标记为转移性组织，则该斑块被标记为阳性。如果一个WSI中至少包含一个阳性斑块，则该WSI被标记为阳性。数据集提供了标准化实验的训练/测试分割。
The CAMELYON16MIL dataset is an adapted version of the CAMELYON16 dataset tailored for Multiple Instance Learning (MIL). This dataset contains preprocessed whole-slide images (WSIs) of lymph node sections, with relevant patches extracted and features computed for each patch. A patch is labeled as positive if more than 50% of its pixels are annotated as metastatic, and a WSI is labeled as positive if it contains at least one positive patch. The dataset provides train/test splits for standardized experimentation.
提供机构：
torchmil
